- Email: [email protected]
Suppression of exercise-induced angina by magnesium sulfate in patients with variant angina
JACC Vol. 12. No. 5 Novombcr W8R:I 177~I!3
Supgresskm
II77
of
Exercise-Induced
Angina by Magnesium Sulfate ian
Patients With Variant Angina KIYOTAKA KUGIYAMA, MD,’ RJr,Z;‘t’?JI YASUE. MD, KEN OKUMURA, MD, KAZUO GQTO, MD, KOTAkO MINODA. MD. HIROO MIYAGi. MD, KOSHI MATSUYAMA. AKIHIRO KOJIMA, MSc, YUKINORI K~rmomoro
KOGA, MD, MUTSUMASA
TAKAHASHI,
MD.
MD
Cily, Jupun
The effects of irr&ravenous magnesiumon
exercise-induced
anginawere examinedin I5 patientswi%variant szgka and in 13 p&ientswith stable effort angina and were compared with those of placebo. exercise and thallium-201
Symptom-limitedbicycle
myocardial
xintigraphy
performed after intravenous administration
were
of 0.27 mm&
kg bodyweight nf -iurn sulfate and after ptaceba on diRerent days. In all patients,serummagnesium levelsafter administrationof magnesiumsulfab were about twofold lighcr than kvek after placebo. Exercise+&ced
segment elevation
angina asseciated with transient ST occurred in 11 patknls
with variant
2 of these patients receiving mngacsiur (p
angina receiving placebo and in only
any patiint withstableeKortangina.In
myocardial ischemia due to .m imbalance between myocardial oxygen demand and supply, and classical stable effort angina has been explained on the basis of increased myocardial oxygen demand in the presence of a fixed coronary stenosis (I ,2). However, short-term reduction of regional myocardial blood flow due to coronary artery spasm has now been accepted as the pathngenetic mechanism for attacks occurring at rest or during exercise in patients with variant
From the Division of Cardiotogy. Department of Radiology. Kumamoto University Medical School. Kumamoto City. lapun. Thi5 work was supported in part by researchgrants for CardiovascularDiseasesfrom the kIpiIfleSe Ministry of Education (A-604401 I) and from the Japanese Minislry of Health and Welfare (6X-l). Manuscript received March 21. 1988: revised mamttcript receivedMay 25, 1988. accepted June 3. 1988. ‘Present address: Section of Cardiology. Haylor College of Medicine. The Methodist Hospital, Houston, Texas 77030. s for re&&: Hirofumi Ynsur. MD. Division of Cardiology. Kumamoto University Medical School. l-l-l Honjo. KumamotoCity. Kumamot0 868. Japan. iullcgc of Cnrdiotugy
administration
of placebo versus after ad!ninistralion
of
magnesium. Thesize uf the perfusiun defect 8~ measured by thallium-201 scintippby was signitkantly lejs in patients
anginareceivingmagnesium thnn that in those receiving placebo ~p
with variant
intravenous magnesium in patients with variant angina but not in patients with stabte etfoft angina. I.rls
ben&cial
e&:1 of magnesium in patients with variant angina is most
due to ilaprovement of regional myocardial b&d flow by suppressiun of coronary artery spasm. (J Am Cdl Cardial 1988;12:3177-8~)
likely
these patients there
Angina pectoris is a clinical syndrome caused by transient
019811 by the Amcriw
was no significant difference in exercise duration after
angina (3-l 1). We and other investigators (612) have shown that the drugof choicefor treatmentof patientswith variant angina is a calcium channel antagonist, which suppresses coronary spasm, rather than a beta-adrenergic blocking agent, which decreases myocardial oxygen demand. Magnesium is one of the most plentiful and important intracellular
cations
in !he body.
Previous
reports
(13.14)
of magi&urn bulLale relieves angina1 symptoms and improves exercise capaci!:l in patients with exertional angina. However. these reports were based on noncontrolled clinical trials, and the mechanism explaining their results remained unclear. Recently (151, WC demonstrated an increased retention rate of loaded magnesium in patients with variant angina or coronary spasm. suggesting that magnesium d&kiency may be involved in the pathogenesis of coronary spasm. Other animal studies (16 19) also support this suggestion on the basis of in vitro experimental findings that magnesium plays a regulatory role in coronary vascular tone by its influence on calcium ion uptake, content and distribution in vascular smooth muscle. Gxxcd
:!ia: pa:cntcrai
use
0735 tw7i8tt/s3..co
KUGIYAMA IX Al. Ef+liCTS OF MAGNI3IUM
1178
ON CORC)NARY
JACC Vol. 12. No. 5 Novcmhcr 19HH:
Ili’Ul3
SPASM
In this context. some studies (20.21) propose that magne-
Table 1. Clinical Characteristics
sium
is nature’s
Variant
there
is no systematic
physiologic
calcium
anlagonisl.
However.
and controlled study concerning the
effects of mugncsium on angina
caused
by coronary
spasm
in
patients with variant angina. We have shown (6-11) that exercise-induced coronary spasm occurs frequently and is a cause of exertional angina with ST segment elevation or depression. or both, in the early morning in many patienls with variant angina. Exercise thalIium-201
myocardial
scintigraphy
provides
magnesium
sulfate
on
1 2 3 4
Age (yr) & Gender 67M 6RM S6M 6IM
quantitative
information concerning myocardial blood flow to the ischemic areas (22.23). Our previous studies (7.8) demonstrated that the technique of combining exercise in the early morning with thallium-201 myocardiai scintigraphy in patients with va;iant angina is useful for the noninvasive evaluation of the effects of antianginal drugs on attacks caused by coronary spasm. In this study, using quantitative rotational thallium-201 tomography. we examined the effects of intravenous
Faticnt No.
exercise-induced
angina
in
different types of patients with angina pectoris.
5 6 7 8 9
10 tt I?
13 14 I5
of IS Patients
With
Angina
46M 49F 71M 58M 68M 72M 49M 4SM 60M S7M 49M
ST Elevation During SpontaneousAttack II. Ill, aVF IL III. aVF V, toV, IL Ill. aVF, and V, lo Vq I. aVL. v, to v, Ill. aVF I. aVL. Vd IO v, II. III. aVF v, II. v, II. II. II.
toV, III. aVF tov3 Ill. aVF III. aVF III. aVF
V, tov,
Coronary Arteriography
Attack W
After NTG
99 99 99 IQ0 99 Ion 90 99 IQ0 100
s2* s2 Sh Sh s3 s5 54 SII SI S6
Normal Normat Normal Normal
loo 99 Km loo loo loo
SI S6 S! s2 54 .%I
Normal 90% S6
Normal Normal Normal Normal Normal
75% 75% 90% 90%
$6 s2 s4 SI
90%
S6
*SI IO SIS indicale segments of the coronaryarteriesshowingspasm.as defined by the American &art AssociationCommittee Report (241. F = female: M = male: NTG = nitroglycerin,
Methods patients. The study population consisted of two different groups of patients with exercise-induced angina. Group I consisted of 15 patients with variant angina (14 men and I woman, aged 45 IO 72 years [mean 581); Group two included I3 patients with stable effort angina (12 men and I woman, aged 44 to 68 years, mean 1571).Group I patients with variant angina fulfilled the following inclusion criteria: I ) .TI)~~II~I~~II.~ rr~trrc.lrs of CIIPSI p&r rrs.wwinted IIVWI ST .wg:nr~w~ ckrwhxz PO.2 mV above the control level) on the electrocardiogram (ECG! at rest. usually in the middle of the night or early morning. and more than two times a day during csclri.\r-in(llrl.rd rrrtgirtd ultarks the sludy. 2) Rcprohri~h associated with ischemic ST segment deviations during rcpcated treadmill or ergometer tests. or both. in the early morning during the study. ST segment deviations during exercise-induced angina appeared in the same ECG leads in which ST segment elevation was observed during the spontaneous attack. 3) Coroonrrr~spcrsn demonstrated arteriographicahy during the angina) altack in a large coronary artery perfusing! the left ventricular area corresponding IO the sirp of ST segment elevation during the spontaneous Study
at&i.:
A
sumtiary
of
the
cliriicai
data
in
patients
with
variant angina i* shown in Table I. Group 2 stul; y patients with stable effort angina had an exercise-induced fixed-threshold angina1 attack with horizontal or downsloping ST segment depression (>O.l mV below the control level), but none had rest or nocturnal angina. ST segment elevation or other clinical findings suggesting coronary spasm. On arteriography. all Group 2 patients had ~70% luminai diameter narrowing in at least
one major coronary artery (four patients in one artery, five in two arteries and four in three arteries). No patient in either group had prior myocardiai infarction, heart failure, impaired renal function or other severe complications. All patients in both groups had normal serum levels of magnesium, potassium and calcium at the time of entry into the study. The two groups of patients did not
significantly differ with respect to age, gender and serum electrolyte levels. Ail patients gave informed consent before they were included in this study. Study protoco!. This study was performed by a singieblind placebo-controlled protocol. In all paiients the study was performed in the early morning because exertional angina occurs most frequently during the first exercise in the morning in patients with variant angina (4-i I), A!! medications were discontinued z-10 days before the study except for nitroglycerin, which was stopped 2 h before the study. In all patients, 0.27 mmollkg body weight of magnesium sulhte (dissolved ir. 100 ml of isotonic glucose) and placebo (100 ml of isotonic glucose) was administered intravenously ovec a period of 20 min early in the morning on the day of infusion. Blood pressure. heart rate and deep tendon reflexes were monitored immediately before and during each infusion. This infusion rate of magnesium sulfate was chosen to cause about a twofold transient elevation of the serum magnesium level at which ievzl serious toxic effects due to hypermagnesemia are unlikely to appear (25). Each infusion of placebo or magnesium was performed singly on separate early morn-
JACC Vol. It. No. 5 November l9lwI 177-
ings with an intervening interval of 1week. The sequence of each tnal was randomized. Exercise and thallium xintigraphic procedure. Using the same methods as described in our previous studies {7,8). each patient performed the exercise and thallium-201 myocardial scintigraphic tests 30 min after completion of each infusion of magnesium sulfate or placebo. The exercise test was performed with the patient in the upright position on an electrically graded bicycle ergometer (380B. Siemens-Elema AB) starting at a work load of 50 W with increments of 25 W every 3 min. A 12 lead electrocardiogram (ECG) and blood pressure (measured by the cuff method) were recorded at rest, during exercise and several minutes after exercise. All patients had undergone the same ergometer test at least twice before the study so that they became familiar with the test procedure. A bolus of 2.5 mCi of thallium-201 was injected through an indwelling intravenous infusion line at the time when anginal chest pain or exhaustion appeared. The patient was encouraged to continue the exercise at the same fevel for 30 s after the injection of thallium-201. lmaging began 5 min after the injection of thallium-201 with use of the same system of single photon emission computed tomography and dedicated computer (ZLC137-ECT and Scintipack 2400, Shimadzu Inc.) described in our previous studies (7,8). The short-axis tomographic images encompassing the entire left ventricle were reconstructed ai 6.0 mm intervals. Quantitativeanalysisof defectsizeon the scintigram. The computerized thallium-201 tomographic method proposed by Garcia et al. (26) and Ritchie et al. (27) was used to quantify the size of the perfusion defect. This approach was also used in our previous studies (7, 8). In brief, circumferential profiles for each short-axis tomographic image were constructed from maximal count values per pixel in each of 60 radii spaced at 6” intervals. Count vs!ues on each point in the profile were then normalized to the maximal count in the profile of each image. The resulting profiles were arranged as a series of concentric circles forming a single twodimensional polar map. Then, extent and severity polar maps were obtained by comparing normalized maximal count values per each point on the generated twodimensional polar map with the corresponding lower normal limits at 2.5 standard deviations below the mean derived
1179
from 30 normal subjects. The extent polar map represents extent of count reduction and !hn .._ ,,.eriiy E.*‘* map represents its sev=;ity Tile erlent score fur the size of Ihr perfzzsiondefect M’NS rlrfn~d by calculating the number of points falting below the corresponding lower normal limits and by expressing this number as a percent of the total left ventricular points on the extent polar map. The severity score was determined by calculating the difference between normalized maximal counts per each point in the area of the defect and corresponding lower normal limits and by dividing this difference by the total number of left ventricular points on the severity map. Magnesium and other electrolytesassay. After the infusions of magnesium sulfate and placebo, blood samples were obtained immediately before the exercise test. Serum magnesium concentration was determined by atomic absorption spectrophotometry (Perkin-Elmer Inc.). Serum calcium and potszclum concentrations .vere determined with a Technicon SMAC-I autoanalyzer. Statistical analysis. Fisher’s exact test was used to compare the incidence of angina and ST segmen! e!ev.&a during the exercise test after magnesium infusion with that after placebo. The serum electrolyte values, hemodynamic variables and scintigraphic defect size were expressed as mean valuek 2 SD and were stadsiically analyzed by use of Student’s phired t test.
Results iutravenausadministrationof magnesiumsulfateand ptacebo (Table 2). In all patients. serum magnesium levels immediately before the exercise test after magnesium sulfate infusion were significantly higher (about twofold) than after placebo. There was no significaut difference in heart rate (67 2 I2 ‘Jersus70 f 12 beatslmin, p = YS) and in systolic blood pr:s:,lre (I30 2 24 versus 12R 5 23 mm Wg. p = NS) between before and at the end of the infusion of magnesium sulfa!e, respectively. Slight cutanecns flushing occurred in all patients during the infusion of magne:;ium su!fate, which promptly disappeured 55 min after termination of the infusion. Toxic effects due to hypermagnesernia, such as depression of respiratory rate and of deep tend& reflexes, were not
IlW
KUGIYAMA
JACC Vol. IZ, No. 5
Ef AL.
Novcmbcr 19811: t 177-63
EFFECTS OF MAGNESIUM ON CORONARY SPASM
Table 3. Exercise Patient Grouts
Variables After
Each Infusion
of Placebo
and Magnesium
Variant Angina (n = 15) Placebo Ex duration Rest HR Rest SBP Peak HR Peak SBP Peak RPP
-_-
3.5 2 672 127 + 103 r 148 + IS.521 +
2.0 IO 26 20 31 5, I65
Magnesium 7.3 * 672 124 2 121 = 173 + 20.995 2
2.1 IS 25 19 32 5.204
sulfate
in Both
Stable Effort Angina (n = 13) Placebo
p Value
5.4 2 75 f 126 f 120 + 164 f 20.019 f
2.9 I3 23 20 26 6.032
Magnesium
p Value
5.0 f 1.7 702 13 130 Lt 21 114 f 19 171 * 23 19.700 f 4.715
NS NS NS NS NS NS
Values are mean + SD. Statistical analysis was performed by use of a Student’s paired I fest. Ex duration = exercise duration (min); HR = hear! rate (beatslmin); RPP = rate-pressure product (mm HgbeaWmin); SBP = systolic blood pressure (mm Hg).
observed in any patient during the infusion of magnesium sulfate. Serum calcium and potassium levels after the administration of magnesium sulfate did not change significantly as compared with levels after the administratioti of placebo. Exercise results in patients with vartant angina (Table 3 and Fig. I to 3). An attack of chest pain was induced during exercise after placebo infusion in ail 15 Group i patients with variant angina, whereas it was induced in only 5 patients after magnesium (p -C 0.001). ST segment elevation occurred transiently after placebo adminisiration during the exerciseinduced attack of chest pain in II of the 15 patients with variant angina whereas it occurred in 2 patients after magnesium (p < 0.005) (Fig. I). The exercise duration in patients with variant angina was longer after magnesium than that after placebo (p < 0.001). Immediately before exercise, the baseline values of heart rate and systolic blood pressure after magnesium infusion were not significandy different from those after placebo infusion. However, at peak exercise,
Figure 1. Case II. Top, Electrocardiogram during the ergometer exercise test in a patient with variant angina whose thailium +cintigrams are shown in Figure 3. Bottom, exercise-induced angina with ST segment elevation was suppressed after intravenous adminislration of magnesium (Mg), BP = blood pressure. REST
both the heart rate and systolic blood pressure were significantly higher after magnesium than those after placebo (heart rate, p < 0.01; systolic blood pressure, p c 0.05). Therefore, the peak rate-pressure product was also significantly greater after magnesium than after placebo {p < 0.01). indicating that myocardial oxygen demand at peak exercise was significantly increased by magnesium infusion. Each area
patient
believed
had
a perfisjon
to be perfirsed
defect
in
the
by the coronary
scinrigraphic artery
in which
extent and severity scores of the scintigraphic perfusion defect were significantly less after magnesium than after placebo (extetit score: 12.7 f 16.9 versus 40.4 f 16.%, p < 0.001; severity score: 0.98 f 1.72 versus 4.49 2 2.59 normalized counts/ point, p < 0.001) (Fig. 2). In all patients with variant angina, the size of the perfusion defect markedly decreased after magnesium ad compared with that after placebo. However,
spasm
was
arteriographically
demonstrated.
The
Figure 2. Effects of magnesium (Mg) on the extent (Mk) and severity (right! scores of defect size on the thallium scintigrams in 15 patients wub variant angina compared with those of placebo. Open circks with horizontal bar indicate mean values. The asterisk indicates five patients with fixed coronary stenosis. 100 r
EXERCISE
Wp
peo.001
ao-
PLACEBO
P l” 8
w-
8
??
40 -8 z k 20 -
OL
??
k
‘t, ??
% 1
PLACEBO
?? I
MO
9
PLACEBO
1
Mg
JACC Vol. 12. No. 5 November 19RR:I 17743
II.
Figure 3. Case Polar representations of myocardial dish-ibution of thallium-201 dunng the ergometer exercise tests in the patient with vanant angina whose efectrocardiogwm during the exercise tests isshown in Figure I. Two-dimensional polar maps !kfi) represent three-dimensional myocardial count distribution on a scale ranging from white (high counts) IO black (low counts). The extent polar maps (center) represent a binary map on which points falling below the corresponding lower normal limits are shown as a black
region and the remaining points fall in a whitz region. The severity polar maps (right) indicate the grade of count reduction at each point on a scale ranging from light gray (low grade) to black (high gqde) in the magnitude of count reduction below lower normal limits. In these polar map representations, activity from the apex of the left ventricle is located at the center. that of
the base at the periphery, inferior waH at the bottom. septum on the left, and lateral wall on the right. The perfusion defect in the anteroseptal area aRer exercise
for those receiving a placebo markedly decreased when administered magnesium. scintigrams after the administration of magnesium. SW imp&ement with magnesium is most likely due to suppression of exercise-induced coronary spasm because the number of patients showing exercise-induced angina with ST segment elevation-which indicates severe reduction of re&al myocardiat blood tlow probably due to coronary spasm (28+29Mecreased significantly after the a$ministration of magnesium. In contrast to the favorable results in patients with variant angina, magnesium did not lessenexercise-induced angina in patients with stable effort angina. This lack of effect may be hue to the failureof magnesium to reduce myocardial oxygen demand, as indirectly demonstrated by the peak ratepressure product. Moreover. magnesium did n& improve myocardial blood flov: to ischemlc areas, as shown by the scintigraphic study. Mechanism d AlsgReJiam e&Is 011coronary spasm. Epidemiologic studies (30-32) suggest that ischemic heart dis-
Figure 4. Effects ofmagnesium (Mg) on the extent Oeftl and severily
Discussion Effect of magnesiumon exerci&nduced spasm.
This
study shows that intravenous
fate suppresses the exercise-induced
angina
and
magnesium
sul-
hi@) scores of thallium defect size on the scintigrams in 13 oatients with stable etlbrt angina compared with those after placebo. Open circles with horizoaltt4 bar indicate mean values. 6r
angina in patients wrth
variant angina. All patients with variant angina performed exercise for a longer period and to a greater work load when receiving magnesium than while receiving placebo, and the peak rate-pressure product, an indirect index of myocardial oxygen demand, increased after magnesium infusion as compared with that after placebo- These beneficial effects in patients with variant angina seem to be due to improvement
of myocardial blood flow to ischemic areas. as demonstrated by the decreased Gte of perfusion defects on the thallium
o-
OI
PLACEBO
I
MI
I
PtACEtJo
I
Mg
1182
KUGIYAMA ET AL. EFFECTS OF MAGNESIUM
JACC Vol. 1.2. No. 5
ON CORONARY
may be related to magnesium intake because the incidence of the disease has been shown to be high in areas with a low concentration of magnesium in drinking water. Furthermore, we have previously observed (15) magnesium deficiency in patients with variant angina. This observation and other reports (16,21,33) suggest that magnesium deficiency may be involved in the pathogenesis of coronary spasm, which has an important role in the many clinical conditions related to ischemic heart disease (3-l I). This line of thinking was supported by the in vitro experimental findings of Turlapaty and Altura (16) that a low extracellular magnesium concentration increased the basal tone of an isolated canine coronary artery and potentiated its contractile responses to circulating vasoconstrictor hormones such as norepinephrine. On the other hand. a high magnesium concentration decreased the basal tone of the coronary artery and depressed its reactivity to contractile agonists. Altura, Turlapaty and other authors (17-W) further showed in vitro evidence that magnesium physiologically competes with calcium, which is essential for regulation of the tone and contraction of vascular smooth muscle at the membrane and at intracellular calcium-binding sites in vascular smooth muscle including the coronary artery, In this sense, magnesium appears to be a natural calcium antagonist (20). Thus, this study in patients with variant angina supports the probability (16-21) that a high magnesium level can inhibit the contraction of coronary vascular smooth muscle. Conclusions. It is now widely accepted that coronary spasm plays an important role in the pathogenesis not only of variant angina but also of other forms of angina ai rest, unstable angina and acute myocardial infarction (3-l I). This stody suggests the possibility that intravenous magnesium may have value as a treatment of these types of ischemic heart disease. ease
We are grateful to Mario S. Verani, MD (Baylor College of Medicine, Houston, Texas) for critical review of the manuscriptand to Maria E. Frias for expert secretarial assistaxe in its preparation. -
References 2. Eps:ctn SE, RcdwoG CR, Goldstein RE, et al. Ar&a pectons. pathcphysiology. evaluation and treatment. Ann Intern Med 1971;75:263-96. 3. Mascri A. Severi S. Nes MD, et al. “Variant” angina: one aspect of a continuous spcttum of vasospaslic myocardial ischemia. Am J Cardiol 1978;42:1019-35. spasm and myocrrdial
7. Kugiyama K. Yasue H, Hurio Y. et al. Effect of propranolol and nifedipine on exercise-induced ~Itack in patients wirh variant angina: assessment by exercise thallium.201 myocardial scinrigraphy with quan+ titative rotalional tomography. Circulation 198&74:37680, 8. Kugiyama K, Yasue H, Okumura K, et al. Simultaneous multivessel coronary artery spasm dtmonstraled by quantitative analysis of thallium201 single ohoton emissioncomputed tomography. Am J. Cardioll987:60: 1009-19. 9. Yasue H. Omote S, Takizawa A. Nagao M, Miwa K, Tanaka S. Circadian variation of exercise capacity in paGents with Prirumetal’s variant angina: mle of exercise-induced coronary arterial spasm. Circulation 1979;59: 938-48. IO. Yasue Ii, Omota S. Takizawa A. Nagao M. Miwa K, Tanaka S. Exertional angina pectoris caused by coronary arterial spasm: effects of various drugs. Am J Cardiol 1979;43:647-52. II.
Yasuc H. Palhophysiology and trcaiment of coronary arterial spasm. Chest t980;78:216.%23S.
12. Schroeder JS, Lamb IH. Bristow MR. Ginsburg R, Hung J. McAuky BJ. Prevention of cardiovascular events in varianl angina by tong-term diltiazem therapy. J Am Coil Cardiol 1983:l:t507-I I. 13. MalkieI-Shapiro B, Bersohn I. Temer PE. Parcnteral magnesium sulfate therapy in coronary hean disease: a preliminary report on its clinical and laboratory aspects. Med Proc 1956;2:455-42. 14. Browne SE. Parenteral magnesium sulfate in arterial disease. Practitioner 1%4:192:791-7. 15. Ooto K, Yasuc H, Kugiyama K, CI ai. Magnesium metabolism in patients with variant angina (abstr). Circula!ion 1987;76&uppt IV):IV-364. 16. Turlapaty PDMV, Altura BM. Magnesium deficiency produces spasms of coronary arteries: relationship to etiology of sudden death in ischcmic heart disease. Science 1980;208:198-200. 17. Attura BM, Akura BT. Influence of magnesium on drug-induced conttaetion and ion content in rabbit aorta. Am J Physiol 1971;220:93844. 18. Turlapaty PDMV, Altura BM. Extracellular magnesium ions control calcium exchange and content of vascutar smooth musck. Eur J Pbarmacol 1978;52:421-3, 19. Hcaly MD. Ward JA. Wang P. Inhibition of calcium by magnesium in the contraction of rat aortic smooth muscle. Magnesium 1984;3:63-72. 20. lscri LT, French JH. Magnesium: nature’s physiologic calcium blocker. Am Heart J 1984,1&3:188-94. 21. Altura BM. Altura BT. Magnesium, ekctrolyte vascular tone. Drugs 198438@uppl i):l-12042.
transport and coronary
22. Nielsen AP, Morris KG. Murdock R, Bruno FP. Cobb FR. Linear relationship between the distribution of thallium-201 and blood flow in ischemic and noninvasive myocardium during exercise. Circulation 1980;61:797-801. 23. Nichols AFJ. Weiss MB, Sciacca RR, Cannon PJ. Blood DK. Relationship between segmental thallium-201 uptake and regional bw flow in patients with coronary artery disease. Circulation 1983;68:310-20. 24. AHA Comhuttee Report, Circulalion 19755t:5-9.
I. Friedberg CK. Angina pcctoris. Circulation 1972;46: 1037-47.
4. Conti CR. Coronary-artery Med 1%33:309.23&9.
November t988:1177-83
SPASM
N Engi J
5. Prinzmetal M. Kennamcr R. Merliss R, Wada T. Bor N. Angina pectoris. I. A variant form of angina pectoris. Am J Med 1959;27:375-88. 6. Yasue H, Omote S, Takizawa A, Nagao M. Coronary arterial spasm in ischcmic heart disease and its palhogencsis: a review. Circ Res I983:52(suppl l):l-147-52.
25. Rude RK. Singer FR. Magnesium deficiency and excess. Ann Rev Med 1981;32:245-59.
et
26. Garcia EV, Train KV, Maddahi J, al. Quantitation of rotational thallium-201 myocardial tomography. J Nucl Med 198526:17-26. 27. Ritchie JL, Davis KB. Williams DL, CatdwelI J. Kennedy JW. Global and regional left functionand tomographic radionuclide perfusion: the Western Washington lntracoronary Streptokinase in Myocardial Infarction Trial. Circuianon 1984:70:867-75.
&entricular
28. Yasue H. Omote S. Takizawa A, et al, Comparison of coronary arteriographic findings during angina pectoris associated with ST elevation or depression. Am J Cardiol 1981;47:53%6. 29. Ekmekci A, Toyoshia H. Kwoczynski JK. Nagaya T. Prinzmetrd M. Angina pectoris. IV. Clinical and experimental difference between ischemia with ST elevation and ischemia with ST depression, Am J Cardiol .I.. _ .._ _. IYbl:I:412-Lb.
JACC Vol. I?,. No. 5 November i%R:t177-83
EFFECTS
T. Crawford MD. Prevalence and pllhological changes of ischemic heart-dicease in a hard-water and in d $oR-water area. Ldncel 1%7;1:229-32.
30. Crawford
31.
Anderson IX’. Neri LC, Schreiber GB, Talt.~vL FDF, Zdrojew~ki lschemic heart disease. waler hardness and myocardial magne+m. Med Assoc J 1975:6(:775-k
32.
OF MAGNESIUM
KUGIYAMA ON COROt4AZY
ET AL. SPASM
1183
Leary WP. Reyes Al. Loekett CJ. Arbuckle DD. Vaq Der Eyl K. Magnewm and Gealh~awribcd lo Ixhemic heart drxa~ in South Africa. s nfr Med I 19l&f#4375-45.
A.
Can
33. Friedman HS. Coronary vasospasm am! its relaGoo4ip deficiency. Magnesium I%2:l:81-3
to magnesium
Supgresskm
II77
of
Exercise-Induced
Angina by Magnesium Sulfate ian
Patients With Variant Angina KIYOTAKA KUGIYAMA, MD,’ RJr,Z;‘t’?JI YASUE. MD, KEN OKUMURA, MD, KAZUO GQTO, MD, KOTAkO MINODA. MD. HIROO MIYAGi. MD, KOSHI MATSUYAMA. AKIHIRO KOJIMA, MSc, YUKINORI K~rmomoro
KOGA, MD, MUTSUMASA
TAKAHASHI,
MD.
MD
Cily, Jupun
The effects of irr&ravenous magnesiumon
exercise-induced
anginawere examinedin I5 patientswi%variant szgka and in 13 p&ientswith stable effort angina and were compared with those of placebo. exercise and thallium-201
Symptom-limitedbicycle
myocardial
xintigraphy
performed after intravenous administration
were
of 0.27 mm&
kg bodyweight nf -iurn sulfate and after ptaceba on diRerent days. In all patients,serummagnesium levelsafter administrationof magnesiumsulfab were about twofold lighcr than kvek after placebo. Exercise+&ced
segment elevation
angina asseciated with transient ST occurred in 11 patknls
with variant
2 of these patients receiving mngacsiur (p
angina receiving placebo and in only
any patiint withstableeKortangina.In
myocardial ischemia due to .m imbalance between myocardial oxygen demand and supply, and classical stable effort angina has been explained on the basis of increased myocardial oxygen demand in the presence of a fixed coronary stenosis (I ,2). However, short-term reduction of regional myocardial blood flow due to coronary artery spasm has now been accepted as the pathngenetic mechanism for attacks occurring at rest or during exercise in patients with variant
From the Division of Cardiotogy. Department of Radiology. Kumamoto University Medical School. Kumamoto City. lapun. Thi5 work was supported in part by researchgrants for CardiovascularDiseasesfrom the kIpiIfleSe Ministry of Education (A-604401 I) and from the Japanese Minislry of Health and Welfare (6X-l). Manuscript received March 21. 1988: revised mamttcript receivedMay 25, 1988. accepted June 3. 1988. ‘Present address: Section of Cardiology. Haylor College of Medicine. The Methodist Hospital, Houston, Texas 77030. s for re&&: Hirofumi Ynsur. MD. Division of Cardiology. Kumamoto University Medical School. l-l-l Honjo. KumamotoCity. Kumamot0 868. Japan. iullcgc of Cnrdiotugy
administration
of placebo versus after ad!ninistralion
of
magnesium. Thesize uf the perfusiun defect 8~ measured by thallium-201 scintippby was signitkantly lejs in patients
anginareceivingmagnesium thnn that in those receiving placebo ~p
with variant
intravenous magnesium in patients with variant angina but not in patients with stabte etfoft angina. I.rls
ben&cial
e&:1 of magnesium in patients with variant angina is most
due to ilaprovement of regional myocardial b&d flow by suppressiun of coronary artery spasm. (J Am Cdl Cardial 1988;12:3177-8~)
likely
these patients there
Angina pectoris is a clinical syndrome caused by transient
019811 by the Amcriw
was no significant difference in exercise duration after
angina (3-l 1). We and other investigators (612) have shown that the drugof choicefor treatmentof patientswith variant angina is a calcium channel antagonist, which suppresses coronary spasm, rather than a beta-adrenergic blocking agent, which decreases myocardial oxygen demand. Magnesium is one of the most plentiful and important intracellular
cations
in !he body.
Previous
reports
(13.14)
of magi&urn bulLale relieves angina1 symptoms and improves exercise capaci!:l in patients with exertional angina. However. these reports were based on noncontrolled clinical trials, and the mechanism explaining their results remained unclear. Recently (151, WC demonstrated an increased retention rate of loaded magnesium in patients with variant angina or coronary spasm. suggesting that magnesium d&kiency may be involved in the pathogenesis of coronary spasm. Other animal studies (16 19) also support this suggestion on the basis of in vitro experimental findings that magnesium plays a regulatory role in coronary vascular tone by its influence on calcium ion uptake, content and distribution in vascular smooth muscle. Gxxcd
:!ia: pa:cntcrai
use
0735 tw7i8tt/s3..co
KUGIYAMA IX Al. Ef+liCTS OF MAGNI3IUM
1178
ON CORC)NARY
JACC Vol. 12. No. 5 Novcmhcr 19HH:
Ili’Ul3
SPASM
In this context. some studies (20.21) propose that magne-
Table 1. Clinical Characteristics
sium
is nature’s
Variant
there
is no systematic
physiologic
calcium
anlagonisl.
However.
and controlled study concerning the
effects of mugncsium on angina
caused
by coronary
spasm
in
patients with variant angina. We have shown (6-11) that exercise-induced coronary spasm occurs frequently and is a cause of exertional angina with ST segment elevation or depression. or both, in the early morning in many patienls with variant angina. Exercise thalIium-201
myocardial
scintigraphy
provides
magnesium
sulfate
on
1 2 3 4
Age (yr) & Gender 67M 6RM S6M 6IM
quantitative
information concerning myocardial blood flow to the ischemic areas (22.23). Our previous studies (7.8) demonstrated that the technique of combining exercise in the early morning with thallium-201 myocardiai scintigraphy in patients with va;iant angina is useful for the noninvasive evaluation of the effects of antianginal drugs on attacks caused by coronary spasm. In this study, using quantitative rotational thallium-201 tomography. we examined the effects of intravenous
Faticnt No.
exercise-induced
angina
in
different types of patients with angina pectoris.
5 6 7 8 9
10 tt I?
13 14 I5
of IS Patients
With
Angina
46M 49F 71M 58M 68M 72M 49M 4SM 60M S7M 49M
ST Elevation During SpontaneousAttack II. Ill, aVF IL III. aVF V, toV, IL Ill. aVF, and V, lo Vq I. aVL. v, to v, Ill. aVF I. aVL. Vd IO v, II. III. aVF v, II. v, II. II. II.
toV, III. aVF tov3 Ill. aVF III. aVF III. aVF
V, tov,
Coronary Arteriography
Attack W
After NTG
99 99 99 IQ0 99 Ion 90 99 IQ0 100
s2* s2 Sh Sh s3 s5 54 SII SI S6
Normal Normat Normal Normal
loo 99 Km loo loo loo
SI S6 S! s2 54 .%I
Normal 90% S6
Normal Normal Normal Normal Normal
75% 75% 90% 90%
$6 s2 s4 SI
90%
S6
*SI IO SIS indicale segments of the coronaryarteriesshowingspasm.as defined by the American &art AssociationCommittee Report (241. F = female: M = male: NTG = nitroglycerin,
Methods patients. The study population consisted of two different groups of patients with exercise-induced angina. Group I consisted of 15 patients with variant angina (14 men and I woman, aged 45 IO 72 years [mean 581); Group two included I3 patients with stable effort angina (12 men and I woman, aged 44 to 68 years, mean 1571).Group I patients with variant angina fulfilled the following inclusion criteria: I ) .TI)~~II~I~~II.~ rr~trrc.lrs of CIIPSI p&r rrs.wwinted IIVWI ST .wg:nr~w~ ckrwhxz PO.2 mV above the control level) on the electrocardiogram (ECG! at rest. usually in the middle of the night or early morning. and more than two times a day during csclri.\r-in(llrl.rd rrrtgirtd ultarks the sludy. 2) Rcprohri~h associated with ischemic ST segment deviations during rcpcated treadmill or ergometer tests. or both. in the early morning during the study. ST segment deviations during exercise-induced angina appeared in the same ECG leads in which ST segment elevation was observed during the spontaneous attack. 3) Coroonrrr~spcrsn demonstrated arteriographicahy during the angina) altack in a large coronary artery perfusing! the left ventricular area corresponding IO the sirp of ST segment elevation during the spontaneous Study
at&i.:
A
sumtiary
of
the
cliriicai
data
in
patients
with
variant angina i* shown in Table I. Group 2 stul; y patients with stable effort angina had an exercise-induced fixed-threshold angina1 attack with horizontal or downsloping ST segment depression (>O.l mV below the control level), but none had rest or nocturnal angina. ST segment elevation or other clinical findings suggesting coronary spasm. On arteriography. all Group 2 patients had ~70% luminai diameter narrowing in at least
one major coronary artery (four patients in one artery, five in two arteries and four in three arteries). No patient in either group had prior myocardiai infarction, heart failure, impaired renal function or other severe complications. All patients in both groups had normal serum levels of magnesium, potassium and calcium at the time of entry into the study. The two groups of patients did not
significantly differ with respect to age, gender and serum electrolyte levels. Ail patients gave informed consent before they were included in this study. Study protoco!. This study was performed by a singieblind placebo-controlled protocol. In all paiients the study was performed in the early morning because exertional angina occurs most frequently during the first exercise in the morning in patients with variant angina (4-i I), A!! medications were discontinued z-10 days before the study except for nitroglycerin, which was stopped 2 h before the study. In all patients, 0.27 mmollkg body weight of magnesium sulhte (dissolved ir. 100 ml of isotonic glucose) and placebo (100 ml of isotonic glucose) was administered intravenously ovec a period of 20 min early in the morning on the day of infusion. Blood pressure. heart rate and deep tendon reflexes were monitored immediately before and during each infusion. This infusion rate of magnesium sulfate was chosen to cause about a twofold transient elevation of the serum magnesium level at which ievzl serious toxic effects due to hypermagnesemia are unlikely to appear (25). Each infusion of placebo or magnesium was performed singly on separate early morn-
JACC Vol. It. No. 5 November l9lwI 177-
ings with an intervening interval of 1week. The sequence of each tnal was randomized. Exercise and thallium xintigraphic procedure. Using the same methods as described in our previous studies {7,8). each patient performed the exercise and thallium-201 myocardial scintigraphic tests 30 min after completion of each infusion of magnesium sulfate or placebo. The exercise test was performed with the patient in the upright position on an electrically graded bicycle ergometer (380B. Siemens-Elema AB) starting at a work load of 50 W with increments of 25 W every 3 min. A 12 lead electrocardiogram (ECG) and blood pressure (measured by the cuff method) were recorded at rest, during exercise and several minutes after exercise. All patients had undergone the same ergometer test at least twice before the study so that they became familiar with the test procedure. A bolus of 2.5 mCi of thallium-201 was injected through an indwelling intravenous infusion line at the time when anginal chest pain or exhaustion appeared. The patient was encouraged to continue the exercise at the same fevel for 30 s after the injection of thallium-201. lmaging began 5 min after the injection of thallium-201 with use of the same system of single photon emission computed tomography and dedicated computer (ZLC137-ECT and Scintipack 2400, Shimadzu Inc.) described in our previous studies (7,8). The short-axis tomographic images encompassing the entire left ventricle were reconstructed ai 6.0 mm intervals. Quantitativeanalysisof defectsizeon the scintigram. The computerized thallium-201 tomographic method proposed by Garcia et al. (26) and Ritchie et al. (27) was used to quantify the size of the perfusion defect. This approach was also used in our previous studies (7, 8). In brief, circumferential profiles for each short-axis tomographic image were constructed from maximal count values per pixel in each of 60 radii spaced at 6” intervals. Count vs!ues on each point in the profile were then normalized to the maximal count in the profile of each image. The resulting profiles were arranged as a series of concentric circles forming a single twodimensional polar map. Then, extent and severity polar maps were obtained by comparing normalized maximal count values per each point on the generated twodimensional polar map with the corresponding lower normal limits at 2.5 standard deviations below the mean derived
1179
from 30 normal subjects. The extent polar map represents extent of count reduction and !hn .._ ,,.eriiy E.*‘* map represents its sev=;ity Tile erlent score fur the size of Ihr perfzzsiondefect M’NS rlrfn~d by calculating the number of points falting below the corresponding lower normal limits and by expressing this number as a percent of the total left ventricular points on the extent polar map. The severity score was determined by calculating the difference between normalized maximal counts per each point in the area of the defect and corresponding lower normal limits and by dividing this difference by the total number of left ventricular points on the severity map. Magnesium and other electrolytesassay. After the infusions of magnesium sulfate and placebo, blood samples were obtained immediately before the exercise test. Serum magnesium concentration was determined by atomic absorption spectrophotometry (Perkin-Elmer Inc.). Serum calcium and potszclum concentrations .vere determined with a Technicon SMAC-I autoanalyzer. Statistical analysis. Fisher’s exact test was used to compare the incidence of angina and ST segmen! e!ev.&a during the exercise test after magnesium infusion with that after placebo. The serum electrolyte values, hemodynamic variables and scintigraphic defect size were expressed as mean valuek 2 SD and were stadsiically analyzed by use of Student’s phired t test.
Results iutravenausadministrationof magnesiumsulfateand ptacebo (Table 2). In all patients. serum magnesium levels immediately before the exercise test after magnesium sulfate infusion were significantly higher (about twofold) than after placebo. There was no significaut difference in heart rate (67 2 I2 ‘Jersus70 f 12 beatslmin, p = YS) and in systolic blood pr:s:,lre (I30 2 24 versus 12R 5 23 mm Wg. p = NS) between before and at the end of the infusion of magnesium sulfa!e, respectively. Slight cutanecns flushing occurred in all patients during the infusion of magne:;ium su!fate, which promptly disappeured 55 min after termination of the infusion. Toxic effects due to hypermagnesernia, such as depression of respiratory rate and of deep tend& reflexes, were not
IlW
KUGIYAMA
JACC Vol. IZ, No. 5
Ef AL.
Novcmbcr 19811: t 177-63
EFFECTS OF MAGNESIUM ON CORONARY SPASM
Table 3. Exercise Patient Grouts
Variables After
Each Infusion
of Placebo
and Magnesium
Variant Angina (n = 15) Placebo Ex duration Rest HR Rest SBP Peak HR Peak SBP Peak RPP
-_-
3.5 2 672 127 + 103 r 148 + IS.521 +
2.0 IO 26 20 31 5, I65
Magnesium 7.3 * 672 124 2 121 = 173 + 20.995 2
2.1 IS 25 19 32 5.204
sulfate
in Both
Stable Effort Angina (n = 13) Placebo
p Value
5.4 2 75 f 126 f 120 + 164 f 20.019 f
2.9 I3 23 20 26 6.032
Magnesium
p Value
5.0 f 1.7 702 13 130 Lt 21 114 f 19 171 * 23 19.700 f 4.715
NS NS NS NS NS NS
Values are mean + SD. Statistical analysis was performed by use of a Student’s paired I fest. Ex duration = exercise duration (min); HR = hear! rate (beatslmin); RPP = rate-pressure product (mm HgbeaWmin); SBP = systolic blood pressure (mm Hg).
observed in any patient during the infusion of magnesium sulfate. Serum calcium and potassium levels after the administration of magnesium sulfate did not change significantly as compared with levels after the administratioti of placebo. Exercise results in patients with vartant angina (Table 3 and Fig. I to 3). An attack of chest pain was induced during exercise after placebo infusion in ail 15 Group i patients with variant angina, whereas it was induced in only 5 patients after magnesium (p -C 0.001). ST segment elevation occurred transiently after placebo adminisiration during the exerciseinduced attack of chest pain in II of the 15 patients with variant angina whereas it occurred in 2 patients after magnesium (p < 0.005) (Fig. I). The exercise duration in patients with variant angina was longer after magnesium than that after placebo (p < 0.001). Immediately before exercise, the baseline values of heart rate and systolic blood pressure after magnesium infusion were not significandy different from those after placebo infusion. However, at peak exercise,
Figure 1. Case II. Top, Electrocardiogram during the ergometer exercise test in a patient with variant angina whose thailium +cintigrams are shown in Figure 3. Bottom, exercise-induced angina with ST segment elevation was suppressed after intravenous adminislration of magnesium (Mg), BP = blood pressure. REST
both the heart rate and systolic blood pressure were significantly higher after magnesium than those after placebo (heart rate, p < 0.01; systolic blood pressure, p c 0.05). Therefore, the peak rate-pressure product was also significantly greater after magnesium than after placebo {p < 0.01). indicating that myocardial oxygen demand at peak exercise was significantly increased by magnesium infusion. Each area
patient
believed
had
a perfisjon
to be perfirsed
defect
in
the
by the coronary
scinrigraphic artery
in which
extent and severity scores of the scintigraphic perfusion defect were significantly less after magnesium than after placebo (extetit score: 12.7 f 16.9 versus 40.4 f 16.%, p < 0.001; severity score: 0.98 f 1.72 versus 4.49 2 2.59 normalized counts/ point, p < 0.001) (Fig. 2). In all patients with variant angina, the size of the perfusion defect markedly decreased after magnesium ad compared with that after placebo. However,
spasm
was
arteriographically
demonstrated.
The
Figure 2. Effects of magnesium (Mg) on the extent (Mk) and severity (right! scores of defect size on the thallium scintigrams in 15 patients wub variant angina compared with those of placebo. Open circks with horizontal bar indicate mean values. The asterisk indicates five patients with fixed coronary stenosis. 100 r
EXERCISE
Wp
peo.001
ao-
PLACEBO
P l” 8
w-
8
??
40 -8 z k 20 -
OL
??
k
‘t, ??
% 1
PLACEBO
?? I
MO
9
PLACEBO
1
Mg
JACC Vol. 12. No. 5 November 19RR:I 17743
II.
Figure 3. Case Polar representations of myocardial dish-ibution of thallium-201 dunng the ergometer exercise tests in the patient with vanant angina whose efectrocardiogwm during the exercise tests isshown in Figure I. Two-dimensional polar maps !kfi) represent three-dimensional myocardial count distribution on a scale ranging from white (high counts) IO black (low counts). The extent polar maps (center) represent a binary map on which points falling below the corresponding lower normal limits are shown as a black
region and the remaining points fall in a whitz region. The severity polar maps (right) indicate the grade of count reduction at each point on a scale ranging from light gray (low grade) to black (high gqde) in the magnitude of count reduction below lower normal limits. In these polar map representations, activity from the apex of the left ventricle is located at the center. that of
the base at the periphery, inferior waH at the bottom. septum on the left, and lateral wall on the right. The perfusion defect in the anteroseptal area aRer exercise
for those receiving a placebo markedly decreased when administered magnesium. scintigrams after the administration of magnesium. SW imp&ement with magnesium is most likely due to suppression of exercise-induced coronary spasm because the number of patients showing exercise-induced angina with ST segment elevation-which indicates severe reduction of re&al myocardiat blood tlow probably due to coronary spasm (28+29Mecreased significantly after the a$ministration of magnesium. In contrast to the favorable results in patients with variant angina, magnesium did not lessenexercise-induced angina in patients with stable effort angina. This lack of effect may be hue to the failureof magnesium to reduce myocardial oxygen demand, as indirectly demonstrated by the peak ratepressure product. Moreover. magnesium did n& improve myocardial blood flov: to ischemlc areas, as shown by the scintigraphic study. Mechanism d AlsgReJiam e&Is 011coronary spasm. Epidemiologic studies (30-32) suggest that ischemic heart dis-
Figure 4. Effects ofmagnesium (Mg) on the extent Oeftl and severily
Discussion Effect of magnesiumon exerci&nduced spasm.
This
study shows that intravenous
fate suppresses the exercise-induced
angina
and
magnesium
sul-
hi@) scores of thallium defect size on the scintigrams in 13 oatients with stable etlbrt angina compared with those after placebo. Open circles with horizoaltt4 bar indicate mean values. 6r
angina in patients wrth
variant angina. All patients with variant angina performed exercise for a longer period and to a greater work load when receiving magnesium than while receiving placebo, and the peak rate-pressure product, an indirect index of myocardial oxygen demand, increased after magnesium infusion as compared with that after placebo- These beneficial effects in patients with variant angina seem to be due to improvement
of myocardial blood flow to ischemic areas. as demonstrated by the decreased Gte of perfusion defects on the thallium
o-
OI
PLACEBO
I
MI
I
PtACEtJo
I
Mg
1182
KUGIYAMA ET AL. EFFECTS OF MAGNESIUM
JACC Vol. 1.2. No. 5
ON CORONARY
may be related to magnesium intake because the incidence of the disease has been shown to be high in areas with a low concentration of magnesium in drinking water. Furthermore, we have previously observed (15) magnesium deficiency in patients with variant angina. This observation and other reports (16,21,33) suggest that magnesium deficiency may be involved in the pathogenesis of coronary spasm, which has an important role in the many clinical conditions related to ischemic heart disease (3-l I). This line of thinking was supported by the in vitro experimental findings of Turlapaty and Altura (16) that a low extracellular magnesium concentration increased the basal tone of an isolated canine coronary artery and potentiated its contractile responses to circulating vasoconstrictor hormones such as norepinephrine. On the other hand. a high magnesium concentration decreased the basal tone of the coronary artery and depressed its reactivity to contractile agonists. Altura, Turlapaty and other authors (17-W) further showed in vitro evidence that magnesium physiologically competes with calcium, which is essential for regulation of the tone and contraction of vascular smooth muscle at the membrane and at intracellular calcium-binding sites in vascular smooth muscle including the coronary artery, In this sense, magnesium appears to be a natural calcium antagonist (20). Thus, this study in patients with variant angina supports the probability (16-21) that a high magnesium level can inhibit the contraction of coronary vascular smooth muscle. Conclusions. It is now widely accepted that coronary spasm plays an important role in the pathogenesis not only of variant angina but also of other forms of angina ai rest, unstable angina and acute myocardial infarction (3-l I). This stody suggests the possibility that intravenous magnesium may have value as a treatment of these types of ischemic heart disease. ease
We are grateful to Mario S. Verani, MD (Baylor College of Medicine, Houston, Texas) for critical review of the manuscriptand to Maria E. Frias for expert secretarial assistaxe in its preparation. -
References 2. Eps:ctn SE, RcdwoG CR, Goldstein RE, et al. Ar&a pectons. pathcphysiology. evaluation and treatment. Ann Intern Med 1971;75:263-96. 3. Mascri A. Severi S. Nes MD, et al. “Variant” angina: one aspect of a continuous spcttum of vasospaslic myocardial ischemia. Am J Cardiol 1978;42:1019-35. spasm and myocrrdial
7. Kugiyama K. Yasue H, Hurio Y. et al. Effect of propranolol and nifedipine on exercise-induced ~Itack in patients wirh variant angina: assessment by exercise thallium.201 myocardial scinrigraphy with quan+ titative rotalional tomography. Circulation 198&74:37680, 8. Kugiyama K, Yasue H, Okumura K, et al. Simultaneous multivessel coronary artery spasm dtmonstraled by quantitative analysis of thallium201 single ohoton emissioncomputed tomography. Am J. Cardioll987:60: 1009-19. 9. Yasue H. Omote S, Takizawa A. Nagao M, Miwa K, Tanaka S. Circadian variation of exercise capacity in paGents with Prirumetal’s variant angina: mle of exercise-induced coronary arterial spasm. Circulation 1979;59: 938-48. IO. Yasue Ii, Omota S. Takizawa A. Nagao M. Miwa K, Tanaka S. Exertional angina pectoris caused by coronary arterial spasm: effects of various drugs. Am J Cardiol 1979;43:647-52. II.
Yasuc H. Palhophysiology and trcaiment of coronary arterial spasm. Chest t980;78:216.%23S.
12. Schroeder JS, Lamb IH. Bristow MR. Ginsburg R, Hung J. McAuky BJ. Prevention of cardiovascular events in varianl angina by tong-term diltiazem therapy. J Am Coil Cardiol 1983:l:t507-I I. 13. MalkieI-Shapiro B, Bersohn I. Temer PE. Parcnteral magnesium sulfate therapy in coronary hean disease: a preliminary report on its clinical and laboratory aspects. Med Proc 1956;2:455-42. 14. Browne SE. Parenteral magnesium sulfate in arterial disease. Practitioner 1%4:192:791-7. 15. Ooto K, Yasuc H, Kugiyama K, CI ai. Magnesium metabolism in patients with variant angina (abstr). Circula!ion 1987;76&uppt IV):IV-364. 16. Turlapaty PDMV, Altura BM. Magnesium deficiency produces spasms of coronary arteries: relationship to etiology of sudden death in ischcmic heart disease. Science 1980;208:198-200. 17. Attura BM, Akura BT. Influence of magnesium on drug-induced conttaetion and ion content in rabbit aorta. Am J Physiol 1971;220:93844. 18. Turlapaty PDMV, Altura BM. Extracellular magnesium ions control calcium exchange and content of vascutar smooth musck. Eur J Pbarmacol 1978;52:421-3, 19. Hcaly MD. Ward JA. Wang P. Inhibition of calcium by magnesium in the contraction of rat aortic smooth muscle. Magnesium 1984;3:63-72. 20. lscri LT, French JH. Magnesium: nature’s physiologic calcium blocker. Am Heart J 1984,1&3:188-94. 21. Altura BM. Altura BT. Magnesium, ekctrolyte vascular tone. Drugs 198438@uppl i):l-12042.
transport and coronary
22. Nielsen AP, Morris KG. Murdock R, Bruno FP. Cobb FR. Linear relationship between the distribution of thallium-201 and blood flow in ischemic and noninvasive myocardium during exercise. Circulation 1980;61:797-801. 23. Nichols AFJ. Weiss MB, Sciacca RR, Cannon PJ. Blood DK. Relationship between segmental thallium-201 uptake and regional bw flow in patients with coronary artery disease. Circulation 1983;68:310-20. 24. AHA Comhuttee Report, Circulalion 19755t:5-9.
I. Friedberg CK. Angina pcctoris. Circulation 1972;46: 1037-47.
4. Conti CR. Coronary-artery Med 1%33:309.23&9.
November t988:1177-83
SPASM
N Engi J
5. Prinzmetal M. Kennamcr R. Merliss R, Wada T. Bor N. Angina pectoris. I. A variant form of angina pectoris. Am J Med 1959;27:375-88. 6. Yasue H, Omote S, Takizawa A, Nagao M. Coronary arterial spasm in ischcmic heart disease and its palhogencsis: a review. Circ Res I983:52(suppl l):l-147-52.
25. Rude RK. Singer FR. Magnesium deficiency and excess. Ann Rev Med 1981;32:245-59.
et
26. Garcia EV, Train KV, Maddahi J, al. Quantitation of rotational thallium-201 myocardial tomography. J Nucl Med 198526:17-26. 27. Ritchie JL, Davis KB. Williams DL, CatdwelI J. Kennedy JW. Global and regional left functionand tomographic radionuclide perfusion: the Western Washington lntracoronary Streptokinase in Myocardial Infarction Trial. Circuianon 1984:70:867-75.
&entricular
28. Yasue H. Omote S. Takizawa A, et al, Comparison of coronary arteriographic findings during angina pectoris associated with ST elevation or depression. Am J Cardiol 1981;47:53%6. 29. Ekmekci A, Toyoshia H. Kwoczynski JK. Nagaya T. Prinzmetrd M. Angina pectoris. IV. Clinical and experimental difference between ischemia with ST elevation and ischemia with ST depression, Am J Cardiol .I.. _ .._ _. IYbl:I:412-Lb.
JACC Vol. I?,. No. 5 November i%R:t177-83
EFFECTS
T. Crawford MD. Prevalence and pllhological changes of ischemic heart-dicease in a hard-water and in d $oR-water area. Ldncel 1%7;1:229-32.
30. Crawford
31.
Anderson IX’. Neri LC, Schreiber GB, Talt.~vL FDF, Zdrojew~ki lschemic heart disease. waler hardness and myocardial magne+m. Med Assoc J 1975:6(:775-k
32.
OF MAGNESIUM
KUGIYAMA ON COROt4AZY
ET AL. SPASM
1183
Leary WP. Reyes Al. Loekett CJ. Arbuckle DD. Vaq Der Eyl K. Magnewm and Gealh~awribcd lo Ixhemic heart drxa~ in South Africa. s nfr Med I 19l&f#4375-45.
A.
Can
33. Friedman HS. Coronary vasospasm am! its relaGoo4ip deficiency. Magnesium I%2:l:81-3
to magnesium